Merge tag 'for-6.12-rc5-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux

Pull btrfs fixes from David Sterba:
 "A few more stability fixes. There's one patch adding export of MIPS
  cmpxchg helper, used in the error propagation fix.

   - fix error propagation from split bios to the original btrfs bio

   - fix merging of adjacent extents (normal operation, defragmentation)

   - fix potential use after free after freeing btrfs device structures"

* tag 'for-6.12-rc5-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux:
  btrfs: fix defrag not merging contiguous extents due to merged extent maps
  btrfs: fix extent map merging not happening for adjacent extents
  btrfs: fix use-after-free of block device file in __btrfs_free_extra_devids()
  btrfs: fix error propagation of split bios
  MIPS: export __cmpxchg_small()

5 files changed, 28 insertions, 30 deletions

diff --git a/fs/btrfs/bio.c b/fs/btrfs/bio.c
index fec5c6cde0a7..7e0f9600b80c 100644
--- a/fs/btrfs/bio.c
+++ b/fs/btrfs/bio.c
@@ -49,6 +49,7 @@ void btrfs_bio_init(struct btrfs_bio *bbio, struct btrfs_fs_info *fs_info,
 	bbio->end_io = end_io;
 	bbio->private = private;
 	atomic_set(&bbio->pending_ios, 1);
+	WRITE_ONCE(bbio->status, BLK_STS_OK);
 }
 
 /*
@@ -113,41 +114,29 @@ static void __btrfs_bio_end_io(struct btrfs_bio *bbio)
 	}
 }
 
-static void btrfs_orig_write_end_io(struct bio *bio);
-
-static void btrfs_bbio_propagate_error(struct btrfs_bio *bbio,
-				       struct btrfs_bio *orig_bbio)
-{
-	/*
-	 * For writes we tolerate nr_mirrors - 1 write failures, so we can't
-	 * just blindly propagate a write failure here.  Instead increment the
-	 * error count in the original I/O context so that it is guaranteed to
-	 * be larger than the error tolerance.
-	 */
-	if (bbio->bio.bi_end_io == &btrfs_orig_write_end_io) {
-		struct btrfs_io_stripe *orig_stripe = orig_bbio->bio.bi_private;
-		struct btrfs_io_context *orig_bioc = orig_stripe->bioc;
-
-		atomic_add(orig_bioc->max_errors, &orig_bioc->error);
-	} else {
-		orig_bbio->bio.bi_status = bbio->bio.bi_status;
-	}
-}
-
 void btrfs_bio_end_io(struct btrfs_bio *bbio, blk_status_t status)
 {
 	bbio->bio.bi_status = status;
 	if (bbio->bio.bi_pool == &btrfs_clone_bioset) {
 		struct btrfs_bio *orig_bbio = bbio->private;
 
-		if (bbio->bio.bi_status)
-			btrfs_bbio_propagate_error(bbio, orig_bbio);
 		btrfs_cleanup_bio(bbio);
 		bbio = orig_bbio;
 	}
 
-	if (atomic_dec_and_test(&bbio->pending_ios))
+	/*
+	 * At this point, bbio always points to the original btrfs_bio. Save
+	 * the first error in it.
+	 */
+	if (status != BLK_STS_OK)
+		cmpxchg(&bbio->status, BLK_STS_OK, status);
+
+	if (atomic_dec_and_test(&bbio->pending_ios)) {
+		/* Load split bio's error which might be set above. */
+		if (status == BLK_STS_OK)
+			bbio->bio.bi_status = READ_ONCE(bbio->status);
 		__btrfs_bio_end_io(bbio);
+	}
 }
 
 static int next_repair_mirror(struct btrfs_failed_bio *fbio, int cur_mirror)
diff --git a/fs/btrfs/bio.h b/fs/btrfs/bio.h
index e48612340745..e2fe16074ad6 100644
--- a/fs/btrfs/bio.h
+++ b/fs/btrfs/bio.h
@@ -79,6 +79,9 @@ struct btrfs_bio {
 	/* File system that this I/O operates on. */
 	struct btrfs_fs_info *fs_info;
 
+	/* Save the first error status of split bio. */
+	blk_status_t status;
+
 	/*
 	 * This member must come last, bio_alloc_bioset will allocate enough
 	 * bytes for entire btrfs_bio but relies on bio being last.
diff --git a/fs/btrfs/defrag.c b/fs/btrfs/defrag.c
index b95ef44c326b..968dae953948 100644
--- a/fs/btrfs/defrag.c
+++ b/fs/btrfs/defrag.c
@@ -763,12 +763,12 @@ static struct extent_map *defrag_lookup_extent(struct inode *inode, u64 start,
 	 * We can get a merged extent, in that case, we need to re-search
 	 * tree to get the original em for defrag.
 	 *
-	 * If @newer_than is 0 or em::generation < newer_than, we can trust
-	 * this em, as either we don't care about the generation, or the
-	 * merged extent map will be rejected anyway.
+	 * This is because even if we have adjacent extents that are contiguous
+	 * and compatible (same type and flags), we still want to defrag them
+	 * so that we use less metadata (extent items in the extent tree and
+	 * file extent items in the inode's subvolume tree).
 	 */
-	if (em && (em->flags & EXTENT_FLAG_MERGED) &&
-	    newer_than && em->generation >= newer_than) {
+	if (em && (em->flags & EXTENT_FLAG_MERGED)) {
 		free_extent_map(em);
 		em = NULL;
 	}
diff --git a/fs/btrfs/extent_map.c b/fs/btrfs/extent_map.c
index 668c617444a5..1d93e1202c33 100644
--- a/fs/btrfs/extent_map.c
+++ b/fs/btrfs/extent_map.c
@@ -230,7 +230,12 @@ static bool mergeable_maps(const struct extent_map *prev, const struct extent_ma
 	if (extent_map_end(prev) != next->start)
 		return false;
 
-	if (prev->flags != next->flags)
+	/*
+	 * The merged flag is not an on-disk flag, it just indicates we had the
+	 * extent maps of 2 (or more) adjacent extents merged, so factor it out.
+	 */
+	if ((prev->flags & ~EXTENT_FLAG_MERGED) !=
+	    (next->flags & ~EXTENT_FLAG_MERGED))
 		return false;
 
 	if (next->disk_bytenr < EXTENT_MAP_LAST_BYTE - 1)
diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c
index 8f340ad1d938..eb51b609190f 100644
--- a/fs/btrfs/volumes.c
+++ b/fs/btrfs/volumes.c
@@ -1105,6 +1105,7 @@ static void btrfs_close_one_device(struct btrfs_device *device)
 	if (device->bdev) {
 		fs_devices->open_devices--;
 		device->bdev = NULL;
+		device->bdev_file = NULL;
 	}
 	clear_bit(BTRFS_DEV_STATE_WRITEABLE, &device->dev_state);
 	btrfs_destroy_dev_zone_info(device);